Draft carbon farming methodology under the EU CRCF falls short of key integrity principles
In April 2026, the European Commission made available a revised draft delegated regulation and Annex setting out a certification methodology for carbon farming activities under the Carbon Removal and Carbon Farming Regulation (CRCF). ‘Carbon farming’ refers to practices resulting in the temporary storage of carbon in natural ecosystems or reduced emissions from soils and includes activities in agriculture and agroforestry on mineral soils, rewetting and restoration of peatlands and of other organic soils, and afforestation. The CRCF is meant to generate high-quality carbon removals and soil emission reductions in the EU. Whether it will be able to do so depends, to a large extent, on the quality of the methodologies that determine which activities are eligible, how mitigation outcomes are quantified, and how integrity risks such as non-additionality, leakage and reversals are addressed.
In this context, we assessed the European Commission’s draft methodology for carbon farming activities which was provided to the CRCF Expert Group in April 2026. We examined whether the proposal adheres to key integrity principles in the voluntary carbon market and compared it to the overarching methodological standards of the PACM. Since PACM methodologies for carbon farming are not yet available, the comparison focuses on the general PACM requirements that apply across activity types, including carbon farming. Our assessment finds that the draft does not adhere to key integrity principles in the voluntary carbon market and sets a lower standard than the Article 6.4 Paris Agreement Crediting Mechanism (PACM). Unless it is significantly revised, many CRCF units from carbon farming are unlikely to be backed by actual emission reductions or removals. If such units were used for compliance with EU climate targets, the CRCF could pose significant risks, potentially undermining the ability of the EU to achieve these targets.
Temporary units without defined use cases
First of all, the draft leaves a major gap regarding the use of CRCF sequestration units. These units expire at the end of the monitoring period unless monitoring is prolonged. Yet the methodology does not define the eligible use cases for such temporary units.
This is a major risk. If temporary units were used to offset permanent emissions, integrity would be undermined and overall emissions could ultimately be higher than without the CRCF. Temporary storage cannot be treated as equivalent to permanent emission reductions. Since units might have been claimed by buyers for offsetting purposes at the time they expire, cancelling expired temporary units from the registry cannot safeguard environmental integrity because it entails no consequences on the offsetting claim that has been made.
CRCF units from carbon farming should therefore not be used for offsetting purposes. Otherwise, the framework creates a direct pathway for low-integrity claims.
Quantification rules leave room for over-crediting
Serious concerns arise with regard to the quantification provisions. Here, the draft methodology leaves too much discretion to operators and several key risks unaddressed.
- First of all, monitoring requirements are not sufficiently operationalised. The methodology does not clearly specify the required monitoring frequency or prescribe robust approaches for key parameters. Requirements for data sources are also insufficient. This is especially problematic for soil carbon measures, where calculated outcomes can diverge substantially from what happens in the field.
- The draft also allows a high degree of flexibility in the choice of quantification approaches. Operators may choose between models, direct measurements and default factors (subject to limited restrictions). The literature has long shown that such flexibility can lead to adverse selection: project participants may select the option that yields the most favourable result rather than the most accurate or conservative one.
- Key uncertainties are likewise not fully accounted for. The methodology fails to address important sources of uncertainty, including uncertainty with regard to business-as-usual assumptions, leakage, associated greenhouse gas emissions, and the use of allometric equations. This makes the outcomes of crediting less conservative and carries the risk that the number of units issued will exceed the actual mitigation.
- Leakage is another major shortcoming of the proposed methodology. The draft does not identify leakage sources, nor does it require their prevention or minimisation; it also does not require any remaining leakage to be quantified. This is particularly problematic for indirect land-use change (ILUC).
- Baseline-setting provisions are also not sufficiently ambitious. Downward adjustments of baselines are only required for agricultural and agroforestry practices that reduce direct and indirect N2O emissions from managed agricultural soils and only from the second crediting period onwards. The methodology does not ensure, across activity types, that baselines become more ambitious over time or continue to reflect changing circumstances.
- Finally, the activity boundary is defined imprecisely and incompletely. Terms such as ‘living biomass’ are used without clear definition, leaving unnecessary room for interpretation. At the same time, important emission sources relevant to carbon farming activities, including fires and soil disruption, are not explicitly included. That is a poor foundation for precise and consistent quantification.
To address these shortcomings, we recommend clearly defining robust and conservative monitoring and quantification requirements; addressing uncertainty comprehensively; including explicit provisions to prevent, minimise and quantify leakage, including ILUC; mandating downward adjustments as well as baseline updates for all activity types; and adding precise definitions of emission sources and carbon pools that must be considered.
Additionality requirements contain important loopholes
While some aspects of the additionality provisions more closely align with PACM requirements, other PACM requirements are not implemented, resulting in important loopholes. Most importantly, operators can register activities under the CRCF even if they have already been operating for several years. In other words, retroactive crediting is allowed. If an activity has already been implemented without the prospect of CRCF units, issuing units for that activity does not generate additional mitigation. The proposed retroactive crediting could create large volumes of non-additional units, and reduce the limited available funding for those activities that truly need them, thereby also undermining the effectiveness and outcomes of the EU CRCF.
The draft also lacks explicit provisions addressing situations where an activity becomes legally required after certification. A further problem is that public and private funding may be combined without separately attributing units to these sources. Without proportional attribution, credited mitigation effects would not be transparently linked to the actual incentive provided by CRCF finance.
We recommend addressing these shortcomings by making the ‘incentive effect test’ mandatory for all operators, excluding activities that are legally required, and proportionally attributing CRCF units when different funding sources are combined.
Permanence risks are not appropriately addressed
Carbon farming activities face non-permanence risks. Carbon stored in soils or biomass can be released again due to management changes, land-use change, or disturbances such as fires or droughts. The methodology should therefore have robust provisions on monitoring, risk assessment, liability and reversals. The current draft does not provide them.
Monitoring is required for only 5 years after the end of the activity period for agricultural/agroforestry activities and for afforestation. For agricultural activities, this means that achieved carbon removals might only be guaranteed for 10 years. While longer monitoring requirements might not be aligned with farmer’s planning horizons, ensuring mitigation impacts for such a short timeframe is not consistent with the objective stated in the CRCF Regulation to store CO₂ “for at least several decades”. Furthermore, for afforestation, high reversal risks remain after the end of the monitoring period but are not accounted for in the draft methodology. Additionally, the methodology lacks incentives for maintaining carbon stocks over longer time periods. We recommend making monitoring mandatory for several decades for afforestation activities.
Furthermore, the draft methodology assumes that no liability mechanism is needed because emission reductions from peatland rewetting are considered irreversible. This is not backed by science. There is a broad scientific consensus that peatland rewetting entails non-permanence risks, even if those risks may be considerably lower than for some other carbon farming activities. Accordingly, major existing carbon crediting programmes such as the UK peatland code and the Verified Carbon Standard (VCS) acknowledge that peatland rewetting has non-permanence risks and require project developers to monitor these and account for any reversal events. Moreover, treating peatland rewetting as effectively permanent fails to address the CRCF Regulation’s requirement of liability mechanisms for soil emission reduction activities. We recommend revising the permanence requirements for peatland rewetting and adding appropriate liability mechanisms for such activities.
The proposed risk assessment for agricultural activities and agroforestry and liability mechanisms for reversals are insufficient. Operators are given considerable flexibility in how they estimate reversal risks, rather than requiring the use of conservative data sources or approaches. The proposed overall risk rate for agricultural activities also appears too low to reflect the actual reversal risks involved. It does not adequately account for accelerated climate change impacts, nor for human-induced reversals. In addition, important liability questions are left unanswered. The draft neither specifies how operators would be held liable for replenishing the buffer pool in the event of avoidable reversals, not does it specify how reversals would be covered if they exceed the credits available in the buffer pool. We recommend accounting for identified risks more comprehensively and strengthening the proposed liability mechanisms in line with best practices on the voluntary carbon market.
Safeguards remain under-specified
The draft methodology also lacks a systematic approach to sustainability safeguards. It does not clearly define the assessment procedure; specific criteria for the different impact categories are not set out; and it does not explain how compliance with the broad sustainability requirements should be demonstrated and verified in practice.
As a result, it remains unclear how operators are expected to monitor impacts, how risks should be managed, and what actions are required if negative impacts are identified. To ensure that safeguards are effective in practice, we recommend providing more detailed guidance on how compliance with sustainability requirements shall be demonstrated by operators.
The low CRCF standard could undermine EU climate action
Most provisions in draft CRCF methodology for carbon farming set a lower standard compared to the PACM and best practice in the voluntary carbon market. The proposed methodology does not draw on the wealth of experience and lessons learned from 20 years of carbon crediting but risks repeating the mistakes of the past that have undermined confidence in carbon credits. Only few requirements align with the PACM, such as parts of the additionality assessment.
A low CRCF standard matters. First of all, this creates large risks of over-crediting and non-additionality across all carbon farming activities. If such units were used for compliance with EU climate targets, the CRCF could pose significant risks, potentially undermining the ability of the EU to achieve these targets.
Secondly, the proposed CRCF carbon farming methodology does to adhere the CRCF Regulation. This creates inconsistency and could potentially pose legal risks for the European Union, because the methodology, or the use of its units, could be challenged in courts. Moreover, Article 18(4) of the CRCF Regulation foresees alignment of CRCF methodologies with Article 6 of the Paris Agreement. The current draft does not achieve that.
Thirdly, the credibility of the EU may be undermined if it is calling for high integrity in international climate negotiations and for the purchase of international credits, while it establishes a domestic carbon crediting mechanism with a low standard. This could contribute to a race to the bottom where non-additional or over-credited units generated under the EU CRCF may outcompete high-integrity credits generated under the PACM or other high-integrity standards. If the EU's crediting framework falls short of international integrity benchmarks, it may not only struggle in terms of environmental credibility, but also be unattractive to buyers who could risk being accused of greenwashing.
It is crucial that the proposed CRCF methodology undergoes substantial revision before its adoption. In revising the methodology, we recommend using the requirements set under the PACM and best practices of other carbon crediting programmes as a benchmark, as well as clarifying the purposes for which CRCF units may be used.
Anne Siemsons is an expert on international climate policy, and she works in the Energy & Climate Division at our Freiburg office.
